Professor Tom Turmezei is a Consultant Musculoskeletal Radiologist at the Norfolk and Norwich University Hospital, Honorary Professor at the University of East Anglia, and Director of KNEE3D. His work includes clinical diagnostic radiology, interventional practice, research, science engagement, publishing and medical innovation, with a particular focus on 3-D imaging in osteoarthritis.
What was your time at Radley like?
I came to Radley in 1991 from the Dragon School in Oxford, where I had grown up. I was really keen to board because I’d seen how much my peers at prep school had enjoyed it. Radley seemed like the perfect option for me – somewhere where both academics and sport were celebrated, in particular rugby. It was all that and more. I made enduring friendships, was challenged intellectually, and was part of the 1995 1st XV that achieved an unbeaten season.
Academically, I always knew I wanted to do medicine. I used to reflect that I should have thought more broadly about career options during my time at Radley (which I was wisely encouraged to do) but looking back now, a career in medicine turned out to be the right choice. Radley prepared me extremely well for university – not just for academic study, but how to work independently, juggle commitments, think critically, and push the boundaries in everything I did.
What was the first part of your journey into medicine like?
I went up to Christ’s College, Cambridge, in 1996 to study Medical Sciences, which was a very science-based course with huge amounts of material to learn. I played a lot of sport alongside working hard and was lucky to represent the University 1st XV on international tours to Japan and Italy. I finished Cambridge with a year of Psychology, as the course allowed a different area of study in the third year. That might sound like a detour, but it ended up being incredibly useful, particularly in understanding human judgement, motivation, communication and how people make decisions.
In 1999, I moved back home to Oxford for Clinical School, where I met my (now) wife who was also training to become a doctor. From there, I qualified, worked as a junior doctor, spent time working in Australia, and gradually realised that while I loved medicine, I didn’t want a career that was fixed purely on clinical practice.
Tom playing at the Prince Chichibu Memorial Stadium in Japan with the Cambridge University Rugby Team in 1998.
When did radiology enter the picture?
Relatively late, actually. I had realised that I didn’t want to become a surgeon, but still wanted something technically demanding, intellectually challenging, and closely linked to anatomy. Radiology brought together a lot of things I cared about and appeared to be reasonably good at: anatomy, technology, pattern recognition and problem-solving. What many people don’t realise, is that it’s also a specialty where you can combine diagnostics with hands-on intervention, still interacting closely with patients, which has been a balance that suits me well.
Your career now looks very different from the stereotypical medical path. When did that begin to happen?
Probably during my specialist training to become a consultant radiologist, when I began to realise how limited some of our diagnostic tools were – particularly in musculoskeletal medicine.
I became interested in osteoarthritis and started asking what felt like a simple question: why are we still relying on 2-D X-ray images to assess a complex 3-D structure like the knee joint? The more I investigated, the more obvious the limitations became.
That led me to a Master’s degree back at Cambridge University, and then, slightly unusually, a PhD at the Cambridge University Engineering Department while continuing some clinical work. I was in my mid-30s by then, asking myself why I was still a student sitting exams. It was a big decision, but it allowed me to develop image analysis techniques that simply didn’t exist before. On the personal side, it also gave me time to spend with my family and make new friends outside of medicine.
What problem were you trying to solve with your PhD?
The core problem lay in trying to find the best possible imaging assessment of osteoarthritis that might replace X-rays. We’re very good at diagnosing late-stage disease with imaging, but much worse at detecting early changes, predicting progression, or tailoring treatment to individuals according to what their disease looks like. And that’s not even considering how genetic and environmental factors also affect these processes.
My PhD focused on developing new 3-D imaging analysis techniques from computed tomography (CT) scans to better understand osteoarthritis; not just whether it’s present, but how it’s likely to evolve, and what that will mean for the patient and their clinical care team. Practically, that involved advanced mathematics, coding, image processing and technical validation. It was challenging, but also hugely satisfying. I feel very grateful to have been supported by the Wellcome Trust, whose funding gave me the opportunity to tackle these problems.
Why is 3-D imaging such a step forward from traditional X-ray images?
Traditional X-ray imaging is cheap, quick and practical, but the image capture flattens a 3-D structure like the knee joint into a 2-D image, losing enormous amounts of information.
With CT, you’re effectively slicing the joint into sections, like cutting a loaf of bread, and then reconstructing it back together virtually; the maths that allows you to do that is fascinating and a reminder of what I learned at Radley as a platform to understand these processes. The 3-D imaging data volume created from the image slices allows you to measure lots of important features such as bone shape and density, joint space width, muscle relationships and much more, with far greater precision than X-ray imaging could ever achieve.
A colourised 3-D map of bone density at the knee joint in someone with a normal, healthy knee. The 3-D structure and bone density information is extracted from a simple CT scan.
In osteoarthritis, this means we can detect subtle changes much earlier, monitor progression more sensitively, and hope to reassure patients that their joints are not getting worse when they would like to strive for a healthy, active lifestyle. That reassurance can be just as powerful as treatment with a disease such as osteoarthritis that can end up being disabling from pain and reduced mobility.
Is that what led to your company KNEE3D?
Absolutely. KNEE3D grew directly out of that research as a different way to have impact. It started as an academic idea: could we create a more sensitive, standardised way of assessing knee joint health that was better than X-ray imaging?
Over time, it became clear that this could also have clinical and commercial applications. KNEE3D now provides 3-D CT imaging analysis services that ultimately aim to help patients understand their knee joint health better and make the right decisions with the help of clinicians and researchers. Think of it like checking in for an MOT of your knee – a way of quantifying each year what’s happening with your knee joint so you feel confident that you can carry on being active, which is actually the frontline strategy for managing early osteoarthritis.
You’ve said that clinical practice is essential for good research. Why?
Because patients teach you things that data never can. You can have the most sophisticated model in the world, but unless you understand how people experience things like pain, fear, and uncertainty, you won’t build something useful. Clinical work keeps you grounded. It forces you to communicate clearly, to be honest about risk, and to understand that everyone is different. It also reminds you that medicine is not just about accuracy – it’s about trust and the desire to help others.
You have also worked with artists and museums, and have been Imaging Editor for Gray’s Anatomy. How did that come about?
That grew out of my love of anatomy and imaging. Artists often see things in the human body that doctors miss, like form, movement, and proportion.
Working with institutions like the Fitzwilliam Museum, Cambridge, and through contacts at the Medical Artists Association and Gray’s Anatomy has been incredibly enriching across everything I do. I believe science benefits from looking at the interaction of aesthetics and technology in different places, whether trying to understand the construction of an ancient Egyptian mummy coffin or tracking the variant course of the ophthalmic artery through the orbit, both of which I have done with the same essential CT imaging.
Visual understanding also matters in medicine and communication, whether you’re diagnosing disease or teaching anatomy. It’s also very human to want to see inside ourselves and the world around us – just look at how many Nobel Prizes have been awarded to revolutionising imaging technologies. I like being part of that mission.
Tom took part in a project where objects from the Fitzwilliam Museum were scanned in facilities at Addenbrooke’s Hospital. CT scanning of Egyptian coffins revealed reuse, hidden histories, and remarkable discoveries, including the world’s youngest known mummified foetus.
With all these different aspects of your work, do you have time for other things?
That’s a good question and something I often get asked. Firstly, I do love my job as a radiologist, so sometimes that doesn’t feel like work. Secondly, my academic life gives me balance with lots of space for thinking differently, also allowing me to travel around the world and make friendships outside of medicine that are enriching and enduring. Finally, I think exploring is hard-wired into my nature, even going so far as to making wine from grapes that I grow myself in Cambridge. That’s a brilliant distraction for obvious reasons. And now I also have two amazing daughters with my wife that still need their dad cheering them on from the sidelines.
Looking back at medicine, what do you think medical imaging will look like in 10 or 20 years?
We’re already moving from purely diagnostic imaging towards predictive imaging. In the future, scans won’t just tell you what’s wrong now, they’ll help estimate what’s likely to happen for individual patients in the future too.
That will involve combining imaging data with machine learning (which includes artificial intelligence), clinical history, genetic and other information such as social factors. Imaging will become just one part of a much larger, integrated health picture.
What kind of people does medicine need now?
Curious people! People who like science, but also like people and enjoy human interaction. Medicine also needs clinicians who can bridge disciplines: who can talk to engineers, data scientists, industry leaders and patients, and translate between them.
It’s also important to emphasize that medicine is far more flexible than people realise. There are roles in research, industry, policy, education, media, and you can move between them. It’s not a failure to change direction; it’s often a strength.
Finally, what would you say to a Radleian who loves science but isn’t sure what to become?
Don’t be put off by the idea that medicine is a constrained career down a single track – it isn’t. It can be a springboard to something else, even another industry. You can explore, travel, adapt and build something that fits who you are.
You do have to work hard, and the training is demanding. But the opportunities are vast, and the satisfaction of helping people, whether directly through clinical care or indirectly through research, is huge. If you’re curious, resilient and interested in the world, medicine will give you room to grow and help find your own place in it.
